IP Network Voice Frame Multiplexing by Codec Delay Category
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In IP-packet supporting networks, transmitting voice frames with varying mouth-to-ear transmission delay budgets requires significant network resources, leading to inefficient bandwidth usage and higher costs due to the need for stricter resource reservations when different codecs with diverse delay requirements are multiplexed together.
Innovation Solution
Attributing codec categories to data frames based on their mouth-to-ear delay budgets, sorting them accordingly, and generating multiplexed cells from frames within the same category to optimize bandwidth usage by adjusting the number of packets per IP cell based on the delay budget of each codec category.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data frames with different delay requirements are multiplexed together in the same IP cell, then bandwidth usage is reduced, but network resource reservation must be stricter leading to less efficient resource utilization
Solution Approach 1:
The patent segments data frames into different categories based on their mouth-to-ear delay budgets. Frames are divided into at least two categories: first category frames with longer tolerable delays and second category frames with shorter tolerable delays. This segmentation allows different QoS treatments for different frame types, resolving the contradiction by enabling efficient multiplexing of compatible frames while protecting delay-sensitive frames.
Solution Approach 2:
The patent applies different quality treatments to different parts of the data stream based on local requirements. Each frame category receives customized handling: first category frames can tolerate longer queuing delays and can be multiplexed more aggressively, while second category frames receive priority treatment with stricter delay guarantees. This local quality differentiation resolves the contradiction by optimizing bandwidth usage for each category according to its specific delay requirements.
2Reliability
If stricter delay bounds are applied to all multiplexed packets, then delay requirements are guaranteed, but resource reservations increase leading to higher costs and less efficient network usage
Solution Approach 1:
The patent segments the data stream into frame categories with different delay budgets. By identifying frames with longer tolerable delays (first category), the system can apply relaxed delay bounds to these frames while maintaining strict bounds only for delay-sensitive frames (second category). This segmentation resolves the contradiction by eliminating the need for universally strict delay bounds, thereby reducing resource reservations while maintaining reliability where needed.
Solution Approach 2:
The patent changes the delay bound parameter dynamically based on frame category. Instead of applying a single strict delay bound to all frames, the system assigns different delay bounds: a first delay bound for first category frames and a second, stricter delay bound for second category frames. This parameter differentiation resolves the contradiction by reducing overall resource reservations while maintaining necessary delay guarantees for sensitive traffic.
Data Source
AI summary
A method for optimizing the use of network resources for the transmission of data signals, such as voice frames, between network units over an IP-packet supporting network. The data frames are obtained using codecs, which may have a different mouth-to-ear transmission delay, from data samples, e.g. voice samples, that are formatted and to which a data frame header is added. The method includes attributing codec categories to each of the data or voice frames according to the codecs by means of which the data samples are generated, each codec category corresponding to a different mouth-to-ear delay budget range for the data frames; sorting the data frames according to their codec categories; generating multiplexed cells from data frames of a same codec category, each multiplexed cell being obtained by multiplex aggregation of a predetermined number of voice samples; and transporting the multiplexed cells from an ingress router to an egress router in the IP-network.